如何为 Postgresql 设计触发器，我的“Hello world”测试有什么问题？

Question

我需要在 postgresql 中使用触发器（最好通过 SQLALchemy ORM）。然而，我似乎无法创建一个基地。 作为基本测试，我尝试在每次插入行时将字符串“Hello world”放入名为 test 的列中。

我这样设置桌子：

from __future__ import annotations
from typing import List
import sqlalchemy
from sqlalchemy import create_engine, ForeignKey, Column, Integer, String, CheckConstraint, MetaData
from sqlalchemy.orm import DeclarativeBase, Mapped, DeclarativeBase, MappedAsDataclass, mapped_column, relationship
from sqlalchemy.schema import DDL
import datetime

dbUrl = 'XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX'

engine = create_engine(dbUrl,
                       echo=True)

metadata = MetaData()



class Base(MappedAsDataclass, DeclarativeBase):
  pass

class Company(Base):
    __tablename__ = "company"
    id: Mapped[int] = mapped_column(primary_key=True)
    name: Mapped[str]
    managers: Mapped[List[Manager]] = relationship(back_populates="company")

Company.__table__
Base.metadata.create_all(engine)

class Employee(Base):
    __tablename__ = "employee"
    id: Mapped[int] = mapped_column(primary_key=True)
    name: Mapped[str]
    type: Mapped[str]
    test: Mapped[str] = mapped_column(nullable=True)

    __mapper_args__ = {
        "polymorphic_identity": "employee",
        "polymorphic_on": "type",
    }

Employee.__table__
Base.metadata.create_all(engine)


class Manager(Employee):
    __tablename__ = "manager"
    id: Mapped[int] = mapped_column(ForeignKey("employee.id"), primary_key=True)
    manager_name: Mapped[str]
    CheckConstraint("manager_name == employee.name", name="check1")

    company_id: Mapped[int] = mapped_column(ForeignKey("company.id"))
    company: Mapped[Company] = relationship(back_populates="managers")

    __mapper_args__ = {
        "polymorphic_identity": "manager",
    }

class Engineer(Employee):
    __tablename__ = "engineer"
    id: Mapped[int] = mapped_column(ForeignKey("employee.id"), primary_key=True)
    engineer_name: Mapped[str]

    company_id: Mapped[int] = mapped_column(ForeignKey("company.id"))
    company: Mapped[Company] = relationship(back_populates="engineers")

    __mapper_args__ = {
        "polymorphic_identity": "engineer",
    }  
Engineer.__table__
Base.metadata.create_all(engine)

然后尝试创建这样的触发器：

sqlalchemy.event.listen(metadata,
                       "before_create",
                       DDL("""CREATE TRIGGER autoCreateEmployee
                       AFTER INSERT
                       ON employee
                       FOR EACH ROW
                       INSERT INTO employee(test)
                       VALUES("Hello world"),
                       0)  
                       """))

表已创建并且没有发出错误，但据我所知，未添加触发器。我还尝试使用 DBeaver 直接在 SQL 中运行测试触发器语句：

CREATE TRIGGER autoCreateEmployee
                       AFTER INSERT
                       ON employee
                       FOR EACH ROW
                       INSERT INTO employee(test)
                       VALUES("Hello world")

这给了我这个错误：

SQL Error [42601]: ERROR: syntax error at or near "INSERT"
  Position: 165

Error position: line: 5 pos: 164

我的触发语句有什么问题？为什么我在通过 SQLAlchemy 运行时看不到错误？

编辑：我试图改编这个例子：

更新：根据评论，我再次查看了文档，并意识到我正在查看的一些示例甚至不是 Postgres。现在我可以运行命令来创建函数和触发器而不会出现错误，但据我所知，触发器存在（因为我无法在不删除的情况下重新创建它），但似乎与它所在的表没有关联应该是开启的。

目前我只是运行原始 SQL 以避免 ORM 复杂化：

drop function hello_world;

CREATE FUNCTION hello_world() RETURNS trigger AS $hello_world$
    BEGIN
        INSERT INTO employee(test) VALUES ("HELLO WORLD!");
        RETURN NEW;
    END;
$hello_world$ LANGUAGE plpgsql;

drop trigger hello on employee;

CREATE TRIGGER hello
    AFTER UPDATE ON employee
    FOR EACH ROW
    EXECUTE FUNCTION hello_world();

Edit2：删除了屏幕截图，因为它是错误的屏幕截图并且基于误解。

现状：

我的功能是这样的：

CREATE FUNCTION hello_world()
RETURNS trigger
LANGUAGE PLPGSQL
AS $hello_world$
    BEGIN
        INSERT INTO employee(test) VALUES ('HELLO WORLD!');
    END;
$hello_world$

我像这样创建触发器：

CREATE TRIGGER hello
    AFTER INSERT ON employee
    FOR EACH ROW
    EXECUTE FUNCTION hello_world();

但是，它尝试创建一个新行，其中仅包含测试列中的测试字符串，而不是创建到我刚刚输入数据的行中。我的印象是

FOR EACH ROW

意味着该函数将在插入的所有行上执行。显然我还缺少一些东西。

如果我跑步：

INSERT INTO employee (name,type) VALUES ('Merv', 'manager');

我明白了

SQL Error [23502]: ERROR: null value in column "name" of relation "employee" violates not-null constraint
  Detail: Failing row contains (31, null, null, HELLO WORLD!).
  Where: SQL statement "INSERT INTO employee(test) VALUES ('HELLO WORLD!')"
PL/pgSQL function hello_world() line 3 at SQL statement

所以现在我需要弄清楚如何使用触发器函数定位刚刚修改的行。

Answer 1

这是我的触发器的工作代码。谢谢大家的耐心等待。

CREATE FUNCTION hello_world()
RETURNS trigger
LANGUAGE PLPGSQL
AS $hello_world$
    BEGIN
        NEW.test := 'HeLLo wErLd';
        RETURN NEW;
    END;
$hello_world$


CREATE TRIGGER hello
    BEFORE INSERT ON employee
    FOR EACH ROW
    EXECUTE FUNCTION hello_world();

Answer 2

可能是因为双引号。一般来说，双引号用于表或字段的名称，单引号用于字符串常量。

试试这个：

sqlalchemy.event.listen(metadata,
                       "before_create",
                       DDL("""CREATE TRIGGER autoCreateEmployee
                       AFTER INSERT
                       ON employee
                       FOR EACH ROW
                       INSERT INTO employee(test)
                       VALUES('Hello world'),
                       0)  
                       """))

还有这个：

CREATE TRIGGER autoCreateEmployee
                       AFTER INSERT
                       ON employee
                       FOR EACH ROW
                       INSERT INTO employee(test)
                       VALUES('Hello world')